Ranking and selection with covariates

Haihui Shen; L. Jeff Hong; Xiaowei Zhang

doi:10.1109/WSC.2017.8247946

Ranking and selection with covariates

Haihui Shen, L. Jeff Hong, Xiaowei Zhang

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

5 Scopus citations

Abstract

We consider a new ranking and selection problem in which the performance of each alternative depends on some observable random covariates. The best alternative is thus not constant but depends on the values of the covariates. Assuming a linear model that relates the mean performance of an alternative and the covariates, we design selection procedures producing policies that represent the best alternative as a function in the covariates. We prove that the selection procedures can provide certain statistical guarantee, which is defined via a nontrivial generalization of the concept of probability of correct selection that is widely used in the conventional ranking and selection setting.

Original language	English (US)
Title of host publication	2017 Winter Simulation Conference, WSC 2017
Editors	Victor Chan
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	2137-2148
Number of pages	12
ISBN (Electronic)	9781538634288
DOIs	https://doi.org/10.1109/WSC.2017.8247946
State	Published - Jun 28 2017
Externally published	Yes
Event	2017 Winter Simulation Conference, WSC 2017 - Las Vegas, United States Duration: Dec 3 2017 → Dec 6 2017

Publication series

Name	Proceedings - Winter Simulation Conference
ISSN (Print)	0891-7736

Conference

Conference	2017 Winter Simulation Conference, WSC 2017
Country/Territory	United States
City	Las Vegas
Period	12/3/17 → 12/6/17

Bibliographical note

Publisher Copyright:
© 2017 IEEE.

Publisher link

10.1109/WSC.2017.8247946

Find It

Find It at U of M Libraries

Cite this

Shen, H, Hong, LJ & Zhang, X 2017, Ranking and selection with covariates. in V Chan (ed.), 2017 Winter Simulation Conference, WSC 2017. Proceedings - Winter Simulation Conference, Institute of Electrical and Electronics Engineers Inc., pp. 2137-2148, 2017 Winter Simulation Conference, WSC 2017, Las Vegas, United States, 12/3/17. https://doi.org/10.1109/WSC.2017.8247946

@inproceedings{db45b124a14e4b72b6db839784743142,

title = "Ranking and selection with covariates",

abstract = "We consider a new ranking and selection problem in which the performance of each alternative depends on some observable random covariates. The best alternative is thus not constant but depends on the values of the covariates. Assuming a linear model that relates the mean performance of an alternative and the covariates, we design selection procedures producing policies that represent the best alternative as a function in the covariates. We prove that the selection procedures can provide certain statistical guarantee, which is defined via a nontrivial generalization of the concept of probability of correct selection that is widely used in the conventional ranking and selection setting.",

author = "Haihui Shen and Hong, {L. Jeff} and Xiaowei Zhang",

note = "Publisher Copyright: {\textcopyright} 2017 IEEE.; 2017 Winter Simulation Conference, WSC 2017 ; Conference date: 03-12-2017 Through 06-12-2017",

year = "2017",

month = jun,

day = "28",

doi = "10.1109/WSC.2017.8247946",

language = "English (US)",

series = "Proceedings - Winter Simulation Conference",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "2137--2148",

editor = "Victor Chan",

booktitle = "2017 Winter Simulation Conference, WSC 2017",

}

TY - GEN

T1 - Ranking and selection with covariates

AU - Shen, Haihui

AU - Hong, L. Jeff

AU - Zhang, Xiaowei

PY - 2017/6/28

Y1 - 2017/6/28

N2 - We consider a new ranking and selection problem in which the performance of each alternative depends on some observable random covariates. The best alternative is thus not constant but depends on the values of the covariates. Assuming a linear model that relates the mean performance of an alternative and the covariates, we design selection procedures producing policies that represent the best alternative as a function in the covariates. We prove that the selection procedures can provide certain statistical guarantee, which is defined via a nontrivial generalization of the concept of probability of correct selection that is widely used in the conventional ranking and selection setting.

AB - We consider a new ranking and selection problem in which the performance of each alternative depends on some observable random covariates. The best alternative is thus not constant but depends on the values of the covariates. Assuming a linear model that relates the mean performance of an alternative and the covariates, we design selection procedures producing policies that represent the best alternative as a function in the covariates. We prove that the selection procedures can provide certain statistical guarantee, which is defined via a nontrivial generalization of the concept of probability of correct selection that is widely used in the conventional ranking and selection setting.

UR - http://www.scopus.com/inward/record.url?scp=85044541872&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85044541872&partnerID=8YFLogxK

U2 - 10.1109/WSC.2017.8247946

DO - 10.1109/WSC.2017.8247946

M3 - Conference contribution

AN - SCOPUS:85044541872

T3 - Proceedings - Winter Simulation Conference

SP - 2137

EP - 2148

BT - 2017 Winter Simulation Conference, WSC 2017

A2 - Chan, Victor

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2017 Winter Simulation Conference, WSC 2017

Y2 - 3 December 2017 through 6 December 2017

ER -

Ranking and selection with covariates

Abstract

Publication series

Conference

Bibliographical note

Publisher link

Find It

Other files and links

Fingerprint

Cite this